A method for obtaining nonparametric estimates of regression models with or without factor-by-curve interactions using local polynomial kernel smoothers or splines. Additionally, a parametric model (allometric model) can be estimated.

`npregfast`

is an R package for obtain nonparametric estimates of regression models
with or without factor-by-curve interactions using local polynomial kernel smoothers or splines.
Additionally, a parametric model (allometric model) can be estimated.
Particular features of the package are facilities for fast smoothness
estimation, and the calculation of their first and second derivative. Users can
define the smoothers parameters. Confidence intervals calculation is provided
by bootstrap methods. Binning techniques were applied to speed up computation
in the estimation and testing processes.

You can view a live interactive demo to see part of its capabilities at http://sestelo.shinyapps.io/npregfast.

`npregfast`

is available through both CRAN and GitHub.

Get the released version from CRAN:

install.packages("npregfast")

Or the development version from GitHub:

# install.packages("devtools")devtools::install_github("sestelo/npregfast")

This file documents software changes since the previous edition.

- the srand function of the fortran code has been deleted. Now it call the R one.

- the random numbers generation (for the bootstrap) has been changed reducing the computational time.
- the plots are now obtained by means of the ggplot2.

- corrected bug at globaltest function (random numbers problem)

- the argument smooth has been incorporated to the functions. Users can select now the type of smoother: kernel (by default) or splines.

- the plotting functions have been changed. Users can choose between the plot.frfast function, used for base graphics, and the autoplot.frfast function, which is appropriate for ggplot2-type plot and returns objects of the ggplot class.
- we have merged the plottdiff function of the previous version of the package with plot.frfast and with autoplot.frfast by means of the new argument diffwith. This new argument lets users visualize the differences between two factor’s levels.